idnits 2.17.1 draft-ietf-v6ops-mech-v2-07.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3667, Section 5.1 on line 14. -- Found old boilerplate from RFC 3978, Section 5.5 on line 1238. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 1215. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 1222. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 1228. ** The document seems to lack an RFC 3978 Section 5.1 IPR Disclosure Acknowledgement -- however, there's a paragraph with a matching beginning. Boilerplate error? ** This document has an original RFC 3978 Section 5.4 Copyright Line, instead of the newer IETF Trust Copyright according to RFC 4748. ** This document has an original RFC 3978 Section 5.5 Disclaimer, instead of the newer disclaimer which includes the IETF Trust according to RFC 4748. ** The document uses RFC 3667 boilerplate or RFC 3978-like boilerplate instead of verbatim RFC 3978 boilerplate. After 6 May 2005, submission of drafts without verbatim RFC 3978 boilerplate is not accepted. The following non-3978 patterns matched text found in the document. That text should be removed or replaced: By submitting this Internet-Draft, I certify that any applicable patent or other IPR claims of which I am aware have been disclosed, or will be disclosed, and any of which I become aware will be disclosed, in accordance with RFC 3668. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- == No 'Intended status' indicated for this document; assuming Proposed Standard == It seems as if not all pages are separated by form feeds - found 0 form feeds but 29 pages Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the RFC 3978 Section 5.4 Copyright Line does not match the current year == The document seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords. (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- Couldn't find a document date in the document -- date freshness check skipped. Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC1122' is defined on line 1024, but no explicit reference was found in the text == Unused Reference: 'RFC1812' is defined on line 1030, but no explicit reference was found in the text == Unused Reference: 'RFC2462' is defined on line 1039, but no explicit reference was found in the text ** Obsolete normative reference: RFC 1981 (Obsoleted by RFC 8201) ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Obsolete normative reference: RFC 2463 (Obsoleted by RFC 4443) == Outdated reference: A later version (-12) exists of draft-ietf-dnsop-ipv6-dns-issues-10 -- Obsolete informational reference (is this intentional?): RFC 2401 (Obsoleted by RFC 4301) -- Obsolete informational reference (is this intentional?): RFC 2461 (Obsoleted by RFC 4861) -- Obsolete informational reference (is this intentional?): RFC 2462 (Obsoleted by RFC 4862) -- Obsolete informational reference (is this intentional?): RFC 2667 (Obsoleted by RFC 4087) -- Obsolete informational reference (is this intentional?): RFC 3484 (Obsoleted by RFC 6724) -- Obsolete informational reference (is this intentional?): RFC 3513 (Obsoleted by RFC 4291) Summary: 8 errors (**), 0 flaws (~~), 8 warnings (==), 13 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 INTERNET-DRAFT E. Nordmark 2 March 29, 2005 Sun Microsystems, Inc. 3 Obsoletes: 2893 R. E. Gilligan 4 Intransa, Inc. 6 Basic Transition Mechanisms for IPv6 Hosts and Routers 7 9 Status of this Memo 11 By submitting this Internet-Draft, I certify that any applicable 12 patent or other IPR claims of which I am aware have been disclosed, 13 and any of which I become aware will be disclosed, in accordance with 14 RFC 3668. 16 Internet-Drafts are working documents of the Internet Engineering 17 Task Force (IETF), its areas, and its working groups. Note that 18 other groups may also distribute working documents as Internet- 19 Drafts. 21 Internet-Drafts are draft documents valid for a maximum of six months 22 and may be updated, replaced, or obsoleted by other documents at any 23 time. It is inappropriate to use Internet-Drafts as reference 24 material or to cite them other than as "work in progress." 26 The list of current Internet-Drafts can be accessed at 27 http://www.ietf.org/ietf/1id-abstracts.txt 29 The list of Internet-Draft Shadow Directories can be accessed at 30 http://www.ietf.org/shadow.html. 32 This draft expires on September 29, 2005. 34 Abstract 36 This document specifies IPv4 compatibility mechanisms that can be 37 implemented by IPv6 hosts and routers. Two mechanisms are specified, 38 "dual stack" and configured tunneling. Dual stack implies providing 39 complete implementations of both versions of the Internet Protocol 40 (IPv4 and IPv6) and configured tunneling provides a means to carry 41 IPv6 packets over unmodified IPv4 routing infrastructures. 43 This document obsoletes RFC 2893. 45 Contents 47 Status of this Memo.......................................... 1 49 1. Introduction............................................. 3 50 1.1. Terminology......................................... 3 52 2. Dual IP Layer Operation.................................. 5 53 2.1. Address Configuration............................... 5 54 2.2. DNS................................................. 5 56 3. Configured Tunneling Mechanisms.......................... 6 57 3.1. Encapsulation....................................... 8 58 3.2. Tunnel MTU and Fragmentation........................ 8 59 3.2.1. Static Tunnel MTU.............................. 9 60 3.2.2. Dynamic Tunnel MTU............................. 10 61 3.3. Hop Limit........................................... 11 62 3.4. Handling ICMPv4 errors.............................. 12 63 3.5. IPv4 Header Construction............................ 14 64 3.6. Decapsulation....................................... 15 65 3.7. Link-Local Addresses................................ 18 66 3.8. Neighbor Discovery over Tunnels..................... 19 68 4. Threat Related to Source Address Spoofing................ 20 70 5. IANA Considerations...................................... 21 72 6. Security Considerations.................................. 21 74 7. Acknowledgments.......................................... 23 76 8. References............................................... 23 77 8.1. Normative References................................ 23 78 8.2. Informative References.............................. 23 80 9. Authors' Addresses....................................... 25 82 10. Changes from RFC 2893................................... 26 84 1. Introduction 86 The key to a successful IPv6 transition is compatibility with the 87 large installed base of IPv4 hosts and routers. Maintaining 88 compatibility with IPv4 while deploying IPv6 will streamline the task 89 of transitioning the Internet to IPv6. This specification defines 90 two mechanisms that IPv6 hosts and routers may implement in order to 91 be compatible with IPv4 hosts and routers. 93 The mechanisms in this document are designed to be employed by IPv6 94 hosts and routers that need to interoperate with IPv4 hosts and 95 utilize IPv4 routing infrastructures. We expect that most nodes in 96 the Internet will need such compatibility for a long time to come, 97 and perhaps even indefinitely. 99 The mechanisms specified here are: 101 - Dual IP layer (also known as Dual Stack): A technique for 102 providing complete support for both Internet protocols -- IPv4 103 and IPv6 -- in hosts and routers. 105 - Configured tunneling of IPv6 over IPv4: A technique for 106 establishing point-to-point tunnels by encapsulating IPv6 107 packets within IPv4 headers to carry them over IPv4 routing 108 infrastructures. 110 The mechanisms defined here are intended to be the core of a 111 "transition toolbox" -- a growing collection of techniques which 112 implementations and users may employ to ease the transition. The 113 tools may be used as needed. Implementations and sites decide which 114 techniques are appropriate to their specific needs. 116 This document defines the basic set of transition mechanisms, but 117 these are not the only tools available. Additional transition and 118 compatibility mechanisms are specified in other documents. 120 1.1. Terminology 122 The following terms are used in this document: 124 Types of Nodes 126 IPv4-only node: 128 A host or router that implements only IPv4. An IPv4- 129 only node does not understand IPv6. The installed base 130 of IPv4 hosts and routers existing before the transition 131 begins are IPv4-only nodes. 133 IPv6/IPv4 node: 135 A host or router that implements both IPv4 and IPv6. 137 IPv6-only node: 139 A host or router that implements IPv6, and does not 140 implement IPv4. The operation of IPv6-only nodes is not 141 addressed in this memo. 143 IPv6 node: 145 Any host or router that implements IPv6. IPv6/IPv4 and 146 IPv6-only nodes are both IPv6 nodes. 148 IPv4 node: 150 Any host or router that implements IPv4. IPv6/IPv4 and 151 IPv4-only nodes are both IPv4 nodes. 153 Techniques Used in the Transition 155 IPv6-over-IPv4 tunneling: 157 The technique of encapsulating IPv6 packets within IPv4 158 so that they can be carried across IPv4 routing 159 infrastructures. 161 Configured tunneling: 163 IPv6-over-IPv4 tunneling where the IPv4 tunnel endpoint 164 address(es) are determined by configuration information 165 on tunnel endpoints. All tunnels are assumed to be 166 bidirectional. The tunnel provides a (virtual) point- 167 to-point link to the IPv6 layer, using the configured 168 IPv4 addresses as the lower layer endpoint addresses. 170 Other transition mechanisms, including other tunneling mechanisms, 171 are outside the scope of this document. 173 The keywords MUST, MUST NOT, REQUIRED, SHALL, SHALL NOT, SHOULD, 174 SHOULD NOT, RECOMMENDED, MAY, and OPTIONAL, when they appear in this 175 document, are to be interpreted as described in [RFC2119]. 177 2. Dual IP Layer Operation 179 The most straightforward way for IPv6 nodes to remain compatible with 180 IPv4-only nodes is by providing a complete IPv4 implementation. IPv6 181 nodes that provide complete IPv4 and IPv6 implementations are called 182 "IPv6/IPv4 nodes." IPv6/IPv4 nodes have the ability to send and 183 receive both IPv4 and IPv6 packets. They can directly interoperate 184 with IPv4 nodes using IPv4 packets, and also directly interoperate 185 with IPv6 nodes using IPv6 packets. 187 Even though a node may be equipped to support both protocols, one or 188 the other stack may be disabled for operational reasons. Here we use 189 a rather loose notion of "stack". A stack being enabled has IP 190 addresses assigned etc, but whether or not any particular application 191 is available on the stacks is explicitly not defined. Thus IPv6/IPv4 192 nodes may be operated in one of three modes: 194 - With their IPv4 stack enabled and their IPv6 stack disabled. 196 - With their IPv6 stack enabled and their IPv4 stack disabled. 198 - With both stacks enabled. 200 IPv6/IPv4 nodes with their IPv6 stack disabled will operate like 201 IPv4-only nodes. Similarly, IPv6/IPv4 nodes with their IPv4 stacks 202 disabled will operate like IPv6-only nodes. IPv6/IPv4 nodes MAY 203 provide a configuration switch to disable either their IPv4 or IPv6 204 stack. 206 The configured tunneling technique, which is described in section 3, 207 may or may not be used in addition to the dual IP layer operation. 209 2.1. Address Configuration 211 Because the nodes support both protocols, IPv6/IPv4 nodes may be 212 configured with both IPv4 and IPv6 addresses. IPv6/IPv4 nodes use 213 IPv4 mechanisms (e.g., DHCP) to acquire their IPv4 addresses, and 214 IPv6 protocol mechanisms (e.g., stateless address autoconfiguration 215 and/or DHCPv6) to acquire their IPv6 addresses. 217 2.2. DNS 219 The Domain Naming System (DNS) is used in both IPv4 and IPv6 to map 220 between hostnames and IP addresses. A new resource record type named 221 "AAAA" has been defined for IPv6 addresses [RFC3596]. Since 222 IPv6/IPv4 nodes must be able to interoperate directly with both IPv4 223 and IPv6 nodes, they must provide resolver libraries capable of 224 dealing with IPv4 "A" records as well as IPv6 "AAAA" records. Note 225 that the lookup of A versus AAAA records is independent of whether 226 the DNS packets are carried in IPv4 or IPv6 packets, and that there 227 is no assumption that the DNS servers know the IPv4/IPv6 capabilities 228 of the requesting node. 230 The issues and operational guidelines for using IPv6 with DNS are 231 described at more length in other documents [DNSOPV6]. 233 DNS resolver libraries on IPv6/IPv4 nodes MUST be capable of handling 234 both AAAA and A records. However, when a query locates an AAAA 235 record holding an IPv6 address, and an A record holding an IPv4 236 address, the resolver library MAY order the results returned to the 237 application in order to influence the version of IP packets used to 238 communicate with that specific node -- IPv6 first, or IPv4 first. 240 The applications SHOULD be able to specify whether they want IPv4, 241 IPv6 or both records [RFC3493]. That defines which address families 242 the resolver looks up. If there isn't an application choice, or if 243 the application has requested both, the resolver library MUST NOT 244 filter out any records. 246 Since most applications try the addresses in the order they are 247 returned by the resolver, this can affect the IP version "preference" 248 of applications. 250 The actual ordering mechanisms are out of scope of this memo. 251 Address selection is described at more length in [RFC3484]. 253 3. Configured Tunneling Mechanisms 255 In most deployment scenarios, the IPv6 routing infrastructure will be 256 built up over time. While the IPv6 infrastructure is being deployed, 257 the existing IPv4 routing infrastructure can remain functional, and 258 can be used to carry IPv6 traffic. Tunneling provides a way to 259 utilize an existing IPv4 routing infrastructure to carry IPv6 260 traffic. 262 IPv6/IPv4 hosts and routers can tunnel IPv6 datagrams over regions of 263 IPv4 routing topology by encapsulating them within IPv4 packets. 264 Tunneling can be used in a variety of ways: 266 - Router-to-Router. IPv6/IPv4 routers interconnected by an IPv4 267 infrastructure can tunnel IPv6 packets between themselves. In 268 this case, the tunnel spans one segment of the end-to-end path 269 that the IPv6 packet takes. 271 - Host-to-Router. IPv6/IPv4 hosts can tunnel IPv6 packets to an 272 intermediary IPv6/IPv4 router that is reachable via an IPv4 273 infrastructure. This type of tunnel spans the first segment of 274 the packet's end-to-end path. 276 - Host-to-Host. IPv6/IPv4 hosts that are interconnected by an 277 IPv4 infrastructure can tunnel IPv6 packets between themselves. 278 In this case, the tunnel spans the entire end-to-end path that 279 the packet takes. 281 - Router-to-Host. IPv6/IPv4 routers can tunnel IPv6 packets to 282 their final destination IPv6/IPv4 host. This tunnel spans only 283 the last segment of the end-to-end path. 285 Configured tunneling can be used in all of the above cases, but is 286 most likely to be used router-to-router due to the need to explicitly 287 configure the tunneling endpoints. 289 The underlying mechanisms for tunneling are: 291 - The entry node of the tunnel (the encapsulator) creates an 292 encapsulating IPv4 header and transmits the encapsulated packet. 294 - The exit node of the tunnel (the decapsulator) receives the 295 encapsulated packet, reassembles the packet if needed, removes 296 the IPv4 header, and processes the received IPv6 packet. 298 - The encapsulator may need to maintain soft state information for 299 each tunnel recording such parameters as the MTU of the tunnel 300 in order to process IPv6 packets forwarded into the tunnel. 302 In configured tunneling, the tunnel endpoint addresses are determined 303 in the encapsulator from configuration information stored for each 304 tunnel. When an IPv6 packet is transmitted over a tunnel, the 305 destination and source addresses for the encapsulating IPv4 header 306 are set as described in Section 3.5. 308 The determination of which packets to tunnel is usually made by 309 routing information on the encapsulator. This is usually done via a 310 routing table, which directs packets based on their destination 311 address using the prefix mask and match technique. 313 The decapsulator matches the received protocol-41 packets to the 314 tunnels it has configured, and allows only the packets where IPv4 315 source addresses match the tunnels configured on the decapsulator. 316 Therefore the operator must ensure that the tunnel's IPv4 address 317 configuration is the same both at the encapsulator and the 318 decapsulator. 320 3.1. Encapsulation 322 The encapsulation of an IPv6 datagram in IPv4 is shown below: 324 +-------------+ 325 | IPv4 | 326 | Header | 327 +-------------+ +-------------+ 328 | IPv6 | | IPv6 | 329 | Header | | Header | 330 +-------------+ +-------------+ 331 | Transport | | Transport | 332 | Layer | ===> | Layer | 333 | Header | | Header | 334 +-------------+ +-------------+ 335 | | | | 336 ~ Data ~ ~ Data ~ 337 | | | | 338 +-------------+ +-------------+ 340 Encapsulating IPv6 in IPv4 342 In addition to adding an IPv4 header, the encapsulator also has to 343 handle some more complex issues: 345 - Determine when to fragment and when to report an ICMPv6 "packet 346 too big" error back to the source. 348 - How to reflect ICMPv4 errors from routers along the tunnel path 349 back to the source as ICMPv6 errors. 351 Those issues are discussed in the following sections. 353 3.2. Tunnel MTU and Fragmentation 355 Naively the encapsulator could view encapsulation as IPv6 using IPv4 356 as a link layer with a very large MTU (65535-20 bytes at most; 20 357 bytes "extra" are needed for the encapsulating IPv4 header). The 358 encapsulator would only need to report ICMPv6 "packet too big" errors 359 back to the source for packets that exceed this MTU. However, such a 360 scheme would be inefficient or non-interoperable for three reasons 361 and therefore MUST NOT be used: 363 1) It would result in more fragmentation than needed. IPv4 layer 364 fragmentation should be avoided due to the performance problems 365 caused by the loss unit being smaller than the retransmission 366 unit [KM97]. 368 2) Any IPv4 fragmentation occurring inside the tunnel, i.e. between 369 the encapsulator and the decapsulator, would have to be 370 reassembled at the tunnel endpoint. For tunnels that terminate 371 at a router, this would require additional memory and other 372 resources to reassemble the IPv4 fragments into a complete IPv6 373 packet before that packet could be forwarded onward. 375 3) The encapsulator has no way of knowing that the decapsulator is 376 able to defragment such IPv4 packets (see Section 3.7 for 377 details), and has no way of knowing that the decapsulator is 378 able to handle such a large IPv6 Maximum Receive Unit (MRU). 380 Hence, the encapsulator MUST NOT treat the tunnel as an interface 381 with an MTU of 64 kilobytes, but instead either use the fixed static 382 MTU or OPTIONAL dynamic MTU determination based on the IPv4 path MTU 383 to the tunnel endpoint. 385 If both the mechanisms are implemented, the decision which to use 386 SHOULD be configurable on a per-tunnel endpoint basis. 388 3.2.1. Static Tunnel MTU 390 A node using static tunnel MTU treats the tunnel interface as having 391 a fixed interface MTU. By default, the MTU MUST be between 1280 and 392 1480 bytes (inclusive), but it SHOULD be 1280 bytes. If the default 393 is not 1280 bytes, the implementation MUST have a configuration knob 394 which can be used to change the MTU value. 396 A node must be able to accept a fragmented IPv6 packet that, after 397 reassembly, is as large as 1500 octets [RFC2460]. This memo also 398 includes requirements (see Section 3.6) for the amount of IPv4 399 reassembly and IPv6 MRU that MUST be supported by all the 400 decapsulators. These ensure correct interoperability with any fixed 401 MTUs between 1280 and 1480 bytes. 403 A larger fixed MTU than supported by these requirements, must not be 404 configured unless it has been administratively ensured that the 405 decapsulator can reassemble or receive packets of that size. 407 The selection of a good tunnel MTU depends on many factors; at least: 409 - Whether the IPv4 protocol-41 packets will be transported over 410 media which may have a lower path MTU (e.g., IPv4 Virtual 411 Private Networks); then picking too high a value might lead to 412 IPv4 fragmentation. 414 - Whether the tunnel is used to transport IPv6 tunneled packets 415 (e.g., a mobile node with an IPv6-in-IPv4 configured tunnel, and 416 an IPv6-in-IPv6 tunnel interface); then picking too low a value 417 might lead to IPv6 fragmentation. 419 If layered encapsulation is believed to be present, it may be prudent 420 to consider supporting dynamic MTU determination instead as it is 421 able to minimize fragmentation and optimize packet sizes. 423 When using the static tunnel MTU the Don't Fragment bit MUST NOT be 424 set in the encapsulating IPv4 header. As a result the encapsulator 425 should not receive any ICMPv4 "packet too big" messages as a result 426 of the packets it has encapsulated. 428 3.2.2. Dynamic Tunnel MTU 430 The dynamic MTU determination is OPTIONAL. However, if it is 431 implemented, it SHOULD have the behavior described in this document. 433 The fragmentation inside the tunnel can be reduced to a minimum by 434 having the encapsulator track the IPv4 Path MTU across the tunnel, 435 using the IPv4 Path MTU Discovery Protocol [RFC1191] and recording 436 the resulting path MTU. The IPv6 layer in the encapsulator can then 437 view a tunnel as a link layer with an MTU equal to the IPv4 path MTU, 438 minus the size of the encapsulating IPv4 header. 440 Note that this does not eliminate IPv4 fragmentation in the case when 441 the IPv4 path MTU would result in an IPv6 MTU less than 1280 bytes. 442 (Any link layer used by IPv6 has to have an MTU of at least 1280 443 bytes [RFC2460].) In this case the IPv6 layer has to "see" a link 444 layer with an MTU of 1280 bytes and the encapsulator has to use IPv4 445 fragmentation in order to forward the 1280 byte IPv6 packets. 447 The encapsulator SHOULD employ the following algorithm to determine 448 when to forward an IPv6 packet that is larger than the tunnel's path 449 MTU using IPv4 fragmentation, and when to return an ICMPv6 "packet 450 too big" message per [RFC1981]: 452 if (IPv4 path MTU - 20) is less than 1280 453 if packet is larger than 1280 bytes 454 Send ICMPv6 "packet too big" with MTU = 1280. 455 Drop packet. 456 else 457 Encapsulate but do not set the Don't Fragment 458 flag in the IPv4 header. The resulting IPv4 459 packet might be fragmented by the IPv4 layer 460 on the encapsulator or by some router along 461 the IPv4 path. 462 endif 463 else 464 if packet is larger than (IPv4 path MTU - 20) 465 Send ICMPv6 "packet too big" with 466 MTU = (IPv4 path MTU - 20). 467 Drop packet. 468 else 469 Encapsulate and set the Don't Fragment flag 470 in the IPv4 header. 471 endif 472 endif 474 Encapsulators that have a large number of tunnels may choose between 475 dynamic versus static tunnel MTU on a per-tunnel endpoint basis. In 476 cases where the number of tunnels that any one node is using is 477 large, it is helpful to observe that this state information can be 478 cached and discarded when not in use. 480 Note that using dynamic tunnel MTU is subject to IPv4 PMTU blackholes 481 should the ICMPv4 "packet too big" messages be dropped by firewalls 482 or not generated by the routers. [RFC1435, RFC2923] 484 3.3. Hop Limit 486 IPv6-over-IPv4 tunnels are modeled as "single-hop" from the IPv6 487 perspective. The tunnel is opaque to users of the network, and is not 488 detectable by network diagnostic tools such as traceroute. 490 The single-hop model is implemented by having the encapsulators and 491 decapsulators process the IPv6 hop limit field as they would if they 492 were forwarding a packet on to any other datalink. That is, they 493 decrement the hop limit by 1 when forwarding an IPv6 packet. (The 494 originating node and final destination do not decrement the hop 495 limit.) 497 The TTL of the encapsulating IPv4 header is selected in an 498 implementation dependent manner. The current suggested value is 499 published in the "Assigned Numbers" RFC [RFC3232][ASSIGNED]. 500 Implementations MAY provide a mechanism to allow the administrator to 501 configure the IPv4 TTL as the IP Tunnel MIB [RFC2667]. 503 3.4. Handling ICMPv4 errors 505 In response to encapsulated packets it has sent into the tunnel, the 506 encapsulator might receive ICMPv4 error messages from IPv4 routers 507 inside the tunnel. These packets are addressed to the encapsulator 508 because it is the IPv4 source of the encapsulated packet. 510 ICMPv4 error handling is only applicable to dynamic MTU 511 determination, even though the functions could be used with static 512 MTU tunnels as well. 514 The ICMPv4 "packet too big" error messages are handled according to 515 IPv4 Path MTU Discovery [RFC1191] and the resulting path MTU is 516 recorded in the IPv4 layer. The recorded path MTU is used by IPv6 to 517 determine if an ICMPv6 "packet too big" error has to be generated as 518 described in section 3.2.2. 520 The handling of other types of ICMPv4 error messages depends on how 521 much information is available from the encapsulated packet that 522 caused the error. 524 Many older IPv4 routers return only 8 bytes of data beyond the IPv4 525 header of the packet in error, which is not enough to include the 526 address fields of the IPv6 header. More modern IPv4 routers are 527 likely to return enough data beyond the IPv4 header to include the 528 entire IPv6 header and possibly even the data beyond that. 530 If sufficient data bytes from the offending packet are available, the 531 encapsulator MAY extract the encapsulated IPv6 packet and use it to 532 generate an ICMPv6 message directed back to the originating IPv6 533 node, as shown below: 535 +--------------+ 536 | IPv4 Header | 537 | dst = encaps | 538 | node | 539 +--------------+ 540 | ICMPv4 | 541 | Header | 542 - - +--------------+ 543 | IPv4 Header | 544 | src = encaps | 545 IPv4 | node | 546 +--------------+ - - 547 Packet | IPv6 | 548 | Header | Original IPv6 549 in +--------------+ Packet - 550 | Transport | Can be used to 551 Error | Header | generate an 552 +--------------+ ICMPv6 553 | | error message 554 ~ Data ~ back to the source. 555 | | 556 - - +--------------+ - - 558 ICMPv4 Error Message Returned to Encapsulating Node 560 When receiving ICMPv4 errors as above and the errors are not "packet 561 too big" it would be useful to log the error as an error related to 562 the tunnel. Also, if sufficient headers are available, then the 563 originating node MAY send an ICMPv6 error of type "unreachable" with 564 code "address unreachable" to the IPv6 source. (The "address 565 unreachable" code is appropriate since, from the perspective of IPv6, 566 the tunnel is a link and that code is used for link-specific errors 567 [RFC2463]). 569 Note that when the IPv4 path MTU is exceeded, and sufficient bytes of 570 payload associated with the ICMPv4 errors are not available, or 571 ICMPv4 errors do not cause the generation of ICMPv6 errors in case 572 there is enough payload, there will be at least two packet drops 573 instead of at least one (the case of a single layer of MTU 574 discovery). Consider a case where an IPv6 host is connected to an 575 IPv4/IPv6 router, which is connected to a network where an ICMPv4 576 error about too big packet size is generated. First the router needs 577 to learn the tunnel (IPv4) MTU which causes at least one packet loss, 578 and then the host needs to learn the (IPv6) MTU from the router which 579 causes at least one packet loss. Still, in all cases there can be 580 more than one packet loss if there are multiple large packets in 581 flight at the same time. 583 3.5. IPv4 Header Construction 585 When encapsulating an IPv6 packet in an IPv4 datagram, the IPv4 586 header fields are set as follows: 588 Version: 590 4 592 IP Header Length in 32-bit words: 594 5 (There are no IPv4 options in the encapsulating 595 header.) 597 Type of Service: 599 0 unless otherwise specified. (See [RFC2983] and 600 [RFC3168] section 9.1 for issues relating to the Type- 601 of-Service byte and tunneling.) 603 Total Length: 605 Payload length from IPv6 header plus length of IPv6 and 606 IPv4 headers (i.e., IPv6 payload length plus a constant 607 60 bytes). 609 Identification: 611 Generated uniquely as for any IPv4 packet transmitted by 612 the system. 614 Flags: 616 Set the Don't Fragment (DF) flag as specified in section 617 3.2. Set the More Fragments (MF) bit as necessary if 618 fragmenting. 620 Fragment offset: 622 Set as necessary if fragmenting. 624 Time to Live: 626 Set in an implementation-specific manner, as described 627 in section 3.3. 629 Protocol: 631 41 (Assigned payload type number for IPv6). 633 Header Checksum: 635 Calculate the checksum of the IPv4 header. [RFC791] 637 Source Address: 639 An IPv4 address of the encapsulator: either configured 640 by the administrator or an address of the outgoing 641 interface. 643 Destination Address: 645 IPv4 address of the tunnel endpoint. 647 When encapsulating the packets, the node must ensure that it will use 648 the correct source address so that the packets are acceptable to the 649 decapsulator as described in Section 3.6. Configuring the source 650 address is appropriate particularly in cases in which automatic 651 selection of source address may produce different results in a 652 certain period of time. This is often the case with multiple 653 addresses, and multiple interfaces, or when routes may change 654 frequently. Therefore, it SHOULD be possible to administratively 655 specify the source address of a tunnel. 657 3.6. Decapsulation 659 When an IPv6/IPv4 host or a router receives an IPv4 datagram that is 660 addressed to one of its own IPv4 addresses or a joined multicast 661 group address, and the value of the protocol field is 41, the packet 662 is potentially a tunnel packet and needs to be verified to belong to 663 one of the configured tunnel interfaces (by checking 664 source/destination addresses), reassembled (if fragmented at the IPv4 665 level), have the IPv4 header removed and the resulting IPv6 datagram 666 be submitted to the IPv6 layer code on the node. 668 The decapsulator MUST verify that the tunnel source address is 669 correct before further processing packets, to mitigate the problems 670 with address spoofing (see section 4). This check also applies to 671 packets which are delivered to transport protocols on the 672 decapsulator. This is done by verifying that the source address is 673 the IPv4 address of the encapsulator, as configured on the 674 decapsulator. Packets for which the IPv4 source address does not 675 match MUST be discarded and an ICMP message SHOULD NOT be generated; 676 however, if the implementation normally sends an ICMP message when 677 receiving an unknown protocol packet, such an error message MAY be 678 sent (e.g., ICMPv4 Protocol 41 Unreachable). 680 A side effect of this address verification is that the node will 681 silently discard packets with a wrong source address, and packets 682 which were received by the node but not directly addressed to it 683 (e.g., broadcast addresses). 685 Independent of any other forms of IPv4 ingress filtering the 686 administrator of the node may have configured, the implementation MAY 687 perform ingress filtering, i.e., check that the packet is arriving 688 from the interface in the direction of the route towards the tunnel 689 end-point, similar to a Strict Reverse Path Forwarding (RPF) check 690 [RFC3704]. As this may cause problems on tunnels which are routed 691 through multiple links, it is RECOMMENDED that this check, if done, 692 is disabled by default. The packets caught by this check SHOULD be 693 discarded; an ICMP message SHOULD NOT be generated by default. 695 The decapsulator MUST be capable of having, on the tunnel interfaces, 696 an IPv6 MRU of at least the maximum of of 1500 bytes and the largest 697 (IPv6) interface MTU on the decapsulator. 699 The decapsulator MUST be capable of reassembling an IPv4 packet that 700 is (after the reassembly) the maximum of 1500 bytes and the largest 701 (IPv4) interface MTU on the decapsulator. The 1500 byte number is a 702 result of encapsulators that use the static MTU scheme in section 703 3.2.1, while encapsulators that use the dynamic scheme in section 704 3.2.2 can cause up to the largest interface MTU on the decapsulator 705 to be received. (Note that it is strictly the interface MTU on the 706 last IPv4 router *before* the decapsulator that matters, but for most 707 links the MTU is the same between all neighbors.) 709 This reassembly limit allows dynamic tunnel MTU determination by the 710 encapsulator to take advantage of larger IPv4 path MTUs. An 711 implementation MAY have a configuration knob which can be used to set 712 a larger value of the tunnel reassembly buffers than the above 713 number, but it MUST NOT be set below the above number. 715 The decapsulation is shown below: 717 +-------------+ 718 | IPv4 | 719 | Header | 720 +-------------+ +-------------+ 721 | IPv6 | | IPv6 | 722 | Header | | Header | 723 +-------------+ +-------------+ 724 | Transport | | Transport | 725 | Layer | ===> | Layer | 726 | Header | | Header | 727 +-------------+ +-------------+ 728 | | | | 729 ~ Data ~ ~ Data ~ 730 | | | | 731 +-------------+ +-------------+ 733 Decapsulating IPv6 from IPv4 735 The decapsulator performs IPv4 reassembly before decapsulating the 736 IPv6 packet. 738 When decapsulating the packet, the IPv6 header is not modified. 739 (However, see [RFC2983] and [RFC3168] section 9.1 for issues relating 740 to the Type of Service byte and tunneling.) If the packet is 741 subsequently forwarded, its hop limit is decremented by one. 743 The encapsulating IPv4 header is discarded, and the resulting packet 744 is checked for validity when submitted to the IPv6 layer. When 745 reconstructing the IPv6 packet the length MUST be determined from the 746 IPv6 payload length since the IPv4 packet might be padded (thus have 747 a length which is larger than the IPv6 packet plus the IPv4 header 748 being removed). 750 After the decapsulation the node MUST silently discard a packet with 751 an invalid IPv6 source address. The list of invalid source addresses 752 SHOULD include at least: 754 - all multicast addresses (FF00::/8) 756 - the loopback address (::1) 758 - all the IPv4-compatible IPv6 addresses [RFC3513] (::/96), 759 excluding the unspecified address for Duplicate Address 760 Detection (::/128) 762 - all the IPv4-mapped IPv6 addresses (::ffff:0:0/96) 764 In addition, the node should be configured to perform ingress 765 filtering [RFC2827][RFC3704] on the IPv6 source address, similar to 766 on any of its interfaces, e.g.: 768 1) if the tunnel is towards the Internet, the node should be 769 configured to check that the site's IPv6 prefixes are not used 770 as the source addresses, or 772 2) if the tunnel is towards an edge network, the node should be 773 configured to check that the source address belongs to that edge 774 network. 776 The prefix lists in the former typically need to be manually 777 configured; the latter could be verified automatically, e.g., by 778 using a strict unicast RPF check, as long as an interface can be 779 designated to be towards an edge. 781 It is RECOMMENDED that the implementations provide a single knob to 782 make it easier to for the administrators to enable strict ingress 783 filtering towards edge networks. 785 3.7. Link-Local Addresses 787 The configured tunnels are IPv6 interfaces (over the IPv4 "link 788 layer") and thus MUST have link-local addresses. The link-local 789 addresses are used by, e.g., routing protocols operating over the 790 tunnels. 792 The interface identifier [RFC3513] for such an interface may be based 793 on the 32-bit IPv4 address of an underlying interface, or formed 794 using some other means, as long as it's unique from the other tunnel 795 endpoint with a reasonably high probability. 797 Note that it may be desirable to form the link-local address in a 798 fashion that minimizes the probability and the effect of having to 799 renumber the link-local address in the event of a topology or 800 hardware change. 802 If an IPv4 address is used for forming the IPv6 link-local address, 803 the interface identifier is the IPv4 address, prepended by zeros. 804 Note that the "Universal/Local" bit is zero, indicating that the 805 interface identifier is not globally unique. The link-local address 806 is formed by appending the interface identifier to the prefix 807 FE80::/64. 809 When the host has more than one IPv4 address in use on the physical 810 interface concerned, a choice of one of these IPv4 addresses is made 811 by the administrator or the implementation when forming the link- 812 local address. 814 +-------+-------+-------+-------+-------+-------+------+------+ 815 | FE 80 00 00 00 00 00 00 | 816 +-------+-------+-------+-------+-------+-------+------+------+ 817 | 00 00 00 00 | IPv4 Address | 818 +-------+-------+-------+-------+-------+-------+------+------+ 820 3.8. Neighbor Discovery over Tunnels 822 Configured tunnel implementations MUST at least accept and respond to 823 the probe packets used by Neighbor Unreachability Detection (NUD) 824 [RFC2461]. The implementations SHOULD also send NUD probe packets to 825 detect when the configured tunnel fails at which point the 826 implementation can use an alternate path to reach the destination. 827 Note that Neighbor Discovery allows that the sending of NUD probes be 828 omitted for router to router links if the routing protocol tracks 829 bidirectional reachability. 831 For the purposes of Neighbor Discovery the configured tunnels 832 specified in this document are assumed to NOT have a link-layer 833 address, even though the link-layer (IPv4) does have an address. 834 This means that: 836 - the sender of Neighbor Discovery packets SHOULD NOT include 837 Source Link Layer Address options or Target Link Layer Address 838 options on the tunnel link. 840 - the receiver MUST, while otherwise processing the Neighbor 841 Discovery packet, silently ignore the content of any Source Link 842 Layer Address options or Target Link Layer Address options 843 received on the tunnel link. 845 Not using a link layer address options is consistent with how 846 Neighbor Discovery is used on other point-to-point links. 848 4. Threat Related to Source Address Spoofing 850 The specification above contains rules that apply tunnel source 851 address verification in particular and ingress filtering 852 [RFC2827][RFC3704] in general to packets before they are 853 decapsulated. When IP-in-IP tunneling (independent of IP versions) 854 is used it is important that this can not be used to bypass any 855 ingress filtering in use for non-tunneled packets. Thus the rules in 856 this document are derived based on should ingress filtering be used 857 for IPv4 and IPv6, the use of tunneling should not provide an easy 858 way to circumvent the filtering. 860 In this case, without specific ingress filtering checks in the 861 decapsulator, it would be possible for an attacker to inject a packet 862 with: 864 - Outer IPv4 source: real IPv4 address of attacker 866 - Outer IPv4 destination: IPv4 address of decapsulator 868 - Inner IPv6 source: Alice which is either the decapsulator or a 869 node close to it. 871 - Inner IPv6 destination: Bob 873 Even if all IPv4 routers between the attacker and the decapsulator 874 implement IPv4 ingress filtering, and all IPv6 routers between the 875 decapsulator and Bob implement IPv6 ingress filtering, the above 876 spoofed packets will not be filtered out. As a result Bob will 877 receive a packet that looks like it was sent from Alice even though 878 the sender was some unrelated node. 880 The solution to this is to have the decapsulator only accept 881 encapsulated packets from the explicitly configured source address 882 (i.e., the other end of the tunnel) as specified in section 3.6. 883 While this does not provide complete protection in the case ingress 884 filtering has not been deployed, it does provide a significant 885 increase in security. The issue and the remainder threats are 886 discussed at more length in Security Considerations. 888 5. IANA Considerations 890 This memo makes no request to IANA. [[ RFC-editor: please remove this 891 section upon publication. ]] 893 6. Security Considerations 895 Generic security considerations of using IPv6 are discussed in a 896 separate document [V6SEC]. 898 An implementation of tunneling needs to be aware that while a tunnel 899 is a link (as defined in [RFC2460]), the threat model for a tunnel 900 might be rather different than for other links, since the tunnel 901 potentially includes all of the Internet. 903 Several mechanisms (e.g., Neighbor Discovery) depend on Hop Count 904 being 255 and/or the addresses being link-local for ensuring that a 905 packet originated on-link, in a semi-trusted environment. Tunnels 906 are more vulnerable to a breach of this assumption than physical 907 links, as an attacker anywhere in the Internet can send an IPv6-in- 908 IPv4 packet to the tunnel decapsulator, causing injection of an 909 encapsulted IPv6 packet to the configured tunnel interface unless the 910 decapsulation checks are able to discard packets injected in such a 911 manner. 913 Therefore, this memo specifies that the decapsulators make these 914 steps (as described in Section 3.6) to mitigate this threat: 916 - IPv4 source address of the packet MUST be the same as configured 917 for the tunnel end-point, 919 - Independent of any IPv4 ingress filtering the administrator may 920 have configured, the implementation MAY perform IPv4 ingress 921 filtering to check that the IPv4 packets are received from an 922 expected interface (but as this may cause some problems, it may 923 be disabled by default), 925 - IPv6 packets with several, obviously invalid IPv6 source 926 addresses received from the tunnel MUST be discarded (see 927 Section 3.6 for details), and 929 - IPv6 ingress filtering should be performed (typically requiring 930 configuration from the operator), to check that the tunneled 931 IPv6 packets are received from an expected interface. 933 Especially the first verification is vital: to avoid this check, the 934 attacker must be able to know the source of the tunnel (ranging from 935 difficult to predictable) and be able to spoof it (easier). 937 If the remainder threats of tunnel source verification are considered 938 to be significant, a tunneling scheme with authentication should be 939 used instead, for example IPsec [RFC2401] (preferable) or Generic 940 Routing Encapsulation with a pre-configured secret key [RFC2890]. As 941 the configured tunnels are set up more or less manually, setting up 942 the keying material is probably not a problem. However, setting up 943 secure IPsec IPv6-in-IPv4 tunnels is described in another document 944 [V64IPSEC]. 946 If the tunneling is done inside an administrative domain, proper 947 ingress filtering at the edge of the domain can also eliminate the 948 threat from outside of the domain. Therefore shorter tunnels are 949 preferable to longer ones, possibly spanning the whole Internet. 951 Additionally, an implementation MUST treat interfaces to different 952 links as separate, e.g., to ensure that Neighbor Discovery packets 953 arriving on one link does not effect other links. This is especially 954 important for tunnel links. 956 When dropping packets due to failing to match the allowed IPv4 source 957 addresses for a tunnel the node should not "acknowledge" the 958 existence of a tunnel, otherwise this could be used to probe the 959 acceptable tunnel endpoint addresses. For that reason, the 960 specification says that such packets MUST be discarded, and an ICMP 961 error message SHOULD NOT be generated, unless the implementation 962 normally sends ICMP destination unreachable messages for unknown 963 protocols; in such a case, the same code MAY be sent. As should be 964 obvious, the not returning the same ICMP code if an error is returned 965 for other protocols may hint that the IPv6 stack (or the protocol 41 966 tunneling processing) has been enabled -- the behaviour should be 967 consistent on how the implementation otherwise behaves to be 968 transparent to probing. 970 7. Acknowledgments 972 We would like to thank the members of the IPv6 working group, the 973 Next Generation Transition (ngtrans) working group, and the v6ops 974 working group for their many contributions and extensive review of 975 this document. Special thanks are due to (in alphabetical order) Jim 976 Bound, Ross Callon, Tim Chown, Alex Conta, Bob Hinden, Bill Manning, 977 John Moy, Mohan Parthasarathy, Chirayu Patel, Pekka Savola, and Fred 978 Templin for many helpful suggestions. Pekka Savola helped in editing 979 the final revisions of the specification. 981 8. References 983 8.1. Normative References 985 [RFC791] J. Postel, "Internet Protocol", RFC 791, September 1981. 987 [RFC1191] Mogul, J., and S. Deering., "Path MTU Discovery", RFC 1191, 988 November 1990. 990 [RFC1981] McCann, J., S. Deering, and J. Mogul. "Path MTU Discovery 991 for IP version 6", RFC 1981, August 1996. 993 [RFC2119] S. Bradner, "Key words for use in RFCs to Indicate 994 Requirement Levels", RFC 2119, March 1997. 996 [RFC2460] Deering, S., and Hinden, R. "Internet Protocol, Version 6 997 (IPv6) Specification", RFC 2460, December 1998. 999 [RFC2463] A. Conta, S. Deering, "Internet Control Message Protocol 1000 (ICMPv6) for the Internet Protocol Version 6 (IPv6) 1001 Specification", RFC 2463, December 1998. 1003 8.2. Informative References 1005 [ASSIGNED] IANA, "Assigned numbers online database", 1006 http://www.iana.org/numbers.html 1008 [DNSOPV6] Durand, A., Ihren, J., and Savola P., "Operational 1009 Considerations and Issues with IPv6 DNS", draft-ietf-dnsop- 1010 ipv6-dns-issues-10.txt, work-in-progress, October 2004. 1012 [KM97] Kent, C., and J. Mogul, "Fragmentation Considered Harmful". 1013 In Proc. SIGCOMM '87 Workshop on Frontiers in Computer 1014 Communications Technology. August 1987. 1016 [V6SEC] P. Savola, "IPv6 Transition/Co-existence Security 1017 Considerations", draft-savola-v6ops-security-overview- 1018 03.txt, work-in-progress, October 2004. 1020 [V64IPSEC] Graveman, R., et al., "Using IPsec to Secure IPv6-over-IPv4 1021 Tunnels", draft-tschofenig-v6ops-secure-tunnels-03.txt, 1022 work-in-progress, December 2004. 1024 [RFC1122] Braden, R., "Requirements for Internet Hosts - Communication 1025 Layers", STD 3, RFC 1122, October 1989. 1027 [RFC1435] S. Knowles, "IESG Advice from Experience with Path MTU 1028 Discovery", RFC 1435, March 1993. 1030 [RFC1812] F. Baker, "Requirements for IP Version 4 Routers", RFC 1812, 1031 June 1995. 1033 [RFC2401] Kent, S., Atkinson, R., "Security Architecture for the 1034 Internet Protocol", RFC 2401, November 1998. 1036 [RFC2461] Narten, T., Nordmark, E., and Simpson, W. "Neighbor 1037 Discovery for IP Version 6 (IPv6)", RFC 2461, December 1998. 1039 [RFC2462] Thomson, S., and Narten, T. "IPv6 Stateless Address 1040 Autoconfiguration," RFC 2462, December 1998. 1042 [RFC2667] D. Thaler, "IP Tunnel MIB", RFC 2667, August 1999. 1044 [RFC2827] Ferguson, P., and Senie, D., "Network Ingress Filtering: 1045 Defeating Denial of Service Attacks which employ IP Source 1046 Address Spoofing", RFC 2827, May 2000. 1048 [RFC2890] Dommety, G., "Key and Sequence Number Extensions to GRE", 1049 RFC 2890, September 2000. 1051 [RFC2923] K. Lahey, "TCP Problems with Path MTU Discovery", RFC 2923, 1052 September 2000. 1054 [RFC2983] D. Black, "Differentiated Services and Tunnels", RFC 2983, 1055 October 2000. 1057 [RFC3056] B. Carpenter, and K. Moore, "Connection of IPv6 Domains via 1058 IPv4 Clouds", RFC 3056, February 2001. 1060 [RFC3168] K. Ramakrishnan, S. Floyd, D. Black, "The Addition of 1061 Explicit Congestion Notification (ECN) to IP", RFC 3168, 1062 September 2001. 1064 [RFC3232] Reynolds, J., "Assigned Numbers: RFC 1700 is Replaced by an 1065 On-line Database", RFC 3232, January 2002. 1067 [RFC3484] R. Draves, "Default Address Selection for IPv6", RFC 3484, 1068 February 2003. 1070 [RFC3493] Gilligan, R., et al, "Basic Socket Interface Extensions for 1071 IPv6", RFC 3493, February 2003. 1073 [RFC3513] Hinden, R., and S. Deering, "IP Version 6 Addressing 1074 Architecture", RFC 3513, April 2003. 1076 [RFC3596] Thomson, S., C. Huitema, V. Ksinant, and M. Souissi, "DNS 1077 Extensions to support IP version 6", RFC 3596, October 2003. 1079 [RFC3704] Baker, F., and Savola P., "Ingress Filtering for Multihomed 1080 Networks", RFC 3704, BCP 84, March 2004. 1082 9. Authors' Addresses 1084 Erik Nordmark 1085 Sun Microsystems Laboratories 1086 180, avenue de l'Europe 1087 38334 SAINT ISMIER Cedex, France 1088 Tel : +33 (0)4 76 18 88 03 1089 Fax : +33 (0)4 76 18 88 88 1090 Email : erik.nordmark@sun.com 1092 Robert E. Gilligan 1093 Intransa, Inc. 1094 2870 Zanker Rd., Suite 100 1095 San Jose, CA 95134 1097 Tel : +1 408 678 8600 1098 Fax : +1 408 678 8800 1099 Email : gilligan@intransa.com, gilligan@leaf.com 1101 10. Changes from RFC 2893 1103 The motivation for the bulk of these changes are to simplify the 1104 document to only contain the mechanisms of wide-spread use. 1106 RFC 2893 contains a mechanism called automatic tunneling. But a much 1107 more general mechanism is specified in RFC 3056 [RFC3056] which gives 1108 each node with a (global) IPv4 address a /48 IPv6 prefix i.e., enough 1109 for a whole site. 1111 The following changes have been performed since RFC 2893: 1113 - Removed references to A6 and retained AAAA. 1115 - Removed automatic tunneling and use of IPv4-compatible 1116 addresses. 1118 - Removed default Configured Tunnel using IPv4 "Anycast Address" 1120 - Removed Source Address Selection section since this is now 1121 covered by another document ([RFC3484]). 1123 - Removed brief mention of 6over4. 1125 - Split into normative and non-normative references and other 1126 reference cleanup. 1128 - Dropped "or equal" in if (IPv4 path MTU - 20) is less than or 1129 equal to 1280 1131 - Dropped this: However, IPv6 may be used in some environments 1132 where interoperability with IPv4 is not required. IPv6 nodes 1133 that are designed to be used in such environments need not use 1134 or even implement these mechanisms. 1136 - Described Static MTU and Dynamic MTU cases separately; clarified 1137 that the dynamic path MTU mechanism is OPTIONAL but if it is 1138 implemented it should follow the rules in section 3.2.2. 1140 - Specified Static MTU to default to a MTU of 1280 to 1480 bytes, 1141 and that this may be configurable. Discussed the issues with 1142 using Static MTU at more length. 1144 - Specified minimal rules for IPv4 reassembly and IPv6 MRU to 1145 enhance interoperability and to minimize blacholes. 1147 - Restated the "currently underway" language about Type-of- 1148 Service, and loosely point at [RFC2983] and [RFC3168]. 1150 - Fixed reference to Assigned Numbers to be to online version 1151 (with proper pointer to "Assigned Numbers is obsolete" RFC). 1153 - Clarified text about ingress filtering e.g. that it applies to 1154 packet delivered to transport protocols on the decapsulator as 1155 well as packets being forwarded by the decapsulator, and how the 1156 decapsulator's checks help when IPv4 and IPv6 ingress filtering 1157 is in place. 1159 - Removed unidirectional tunneling; assume all tunnels are 1160 bidirectional, between endpoint addresses (not nodes). 1162 - Removed the guidelines for advertising addresses in DNS as 1163 slightly out of scope, referring to another document for the 1164 details. 1166 - Removed the SHOULD requirement that the link-local addresses 1167 should be formed based on IPv4 addresses. 1169 - Added a SHOULD for implementing a knob to be able to set the 1170 source address of the tunnel, and add discussion why this is 1171 useful. 1173 - Added stronger wording for source address checks: both IPv4 and 1174 IPv6 source addresses MUST be checked, and RPF-like ingress 1175 filtering is optional. 1177 - Rewrote security considerations to be more precise about the 1178 threats of tunneling. 1180 - Added a note about considering using TTL=255 when encapsulating. 1182 - Added more discussion in Section 3.2 why using an "infinite" 1183 IPv6 MTU leads to likely interoperability problems. 1185 - Added an explicit requirement that if both MTU determination 1186 methods are used, choosing one should be possible on a per- 1187 tunnel basis. 1189 - Clarified that ICMPv4 error handling is only applicable to 1190 dynamic MTU determination. 1192 - Removed/clarified DNS record filtering; an API is a SHOULD and 1193 if it does not exist, MUST NOT filter anything. Decree ordering 1194 out of scope, but refer to RFC3484. 1196 - Add a note that the destination IPv4 address could also be a 1197 multicast address. 1199 - Make it RECOMMENDED to provide a toggle to perform strict 1200 ingress filtering on an interface. 1202 - Generalize the text on the data in ICMPv4 messages. 1204 - Made a lot of miscellaneous editorial cleanups. 1206 Intellectual Property Statement 1208 The IETF takes no position regarding the validity or scope of any 1209 Intellectual Property Rights or other rights that might be claimed to 1210 pertain to the implementation or use of the technology described in 1211 this document or the extent to which any license under such rights 1212 might or might not be available; nor does it represent that it has 1213 made any independent effort to identify any such rights. Information 1214 on the procedures with respect to rights in RFC documents can be 1215 found in BCP 78 and BCP 79. 1217 Copies of IPR disclosures made to the IETF Secretariat and any 1218 assurances of licenses to be made available, or the result of an 1219 attempt made to obtain a general license or permission for the use of 1220 such proprietary rights by implementers or users of this 1221 specification can be obtained from the IETF on-line IPR repository at 1222 http://www.ietf.org/ipr. 1224 The IETF invites any interested party to bring to its attention any 1225 copyrights, patents or patent applications, or other proprietary 1226 rights that may cover technology that may be required to implement 1227 this standard. Please address the information to the IETF at ietf- 1228 ipr@ietf.org. 1230 Disclaimer of Validity 1232 This document and the information contained herein are provided on an 1233 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 1234 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET 1235 ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, 1236 INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE 1237 INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 1238 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 1240 Copyright Statement 1242 Copyright (C) The Internet Society (2004). This document is subject 1243 to the rights, licenses and restrictions contained in BCP 78, and 1244 except as set forth therein, the authors retain all their rights. 1246 Acknowledgment 1248 Funding for the RFC Editor function is currently provided by the 1249 Internet Society.